5 April 2001 Gamma2001 Meeting 1

RXTE Observations of the Vela Pulsar:RXTE Observations of the Vela Pulsar:The Pulsar Rosetta Stone The Pulsar Rosetta Stone

M.S. Strickman NRLA.K. Harding GSFCC. Gwinn UCSBP. McCulloch University of TasmaniaD. Moffett Furman University

5 April 2001 Gamma2001 Meeting 2

PSR B0833PSR B0833--45: The Vela Pulsar45: The Vela Pulsar

• Brightest celestial γ-ray source but difficult X-ray object– Faint– Embedded in bright pulsar wind nebula

• First X-ray observation by ROSAT– Pulsed spectrum described by black body– “Point source” spectrum described by black body or

power law– Pulsed fraction ~11% (<1.2 keV)

5 April 2001 Gamma2001 Meeting 3

Vela Pulsar (cont’d)Vela Pulsar (cont’d)

• First convincing hard X-ray/soft γ-rayobservations by OSSE and COMPTEL– Pulsed spectrum only– Gradual rollover from

very hard OSSE spectrum (Γ =1.3) to EGRET (Γ = 1.7) 10-4 10-2 100 102 104 106

Energy (keV)

10-8

10-6

10-4

10-2

Mea

n B

and

En

ergy

x P

hoto

n F

lux

(keV

/cm

2-s

-keV

)

OSSECOMPTELEGRETROSAT Pulsed BlackbodyOptical

But what happens in

here?

5 April 2001 Gamma2001 Meeting 4

Vela Pulsar (cont’d)Vela Pulsar (cont’d)

• Light curve behavior seems to have three “regimes”– Radio: 1 peak– Optical: 2 peaks, small

separation– γ-ray: 2 peaks, wider

separation• Where does the switch

from “optical-like” to “γ-like” behavior occur?

• Whither the radio pulse?

EGRET> 100 MeV

Pk 1 Pk 2

COMPTEL10 - 30 MeV

OSSE0.07 - 0.57 MeV

ROSAT0.06 - 2.4 keV

0.0 0.5 1.0 1.5 2.0Phase

Optical

5 April 2001 Gamma2001 Meeting 5

RXTE Arrives with Answers and QuestionsRXTE Arrives with Answers and Questions

• RXTE Observations– Cycle 1: 93 ksec– Cycle 3: 274 ksec

• Analysis– Ephemerides from

Princeton pulsar database– Epoch folding analysis with

standard RXTE ftools– “On-pulse minus off-pulse”

or fitting sinusoid model plus off-pulse “background”

• Cycle 1 results appeared in 1999 (Strickman, Harding & deJager, 1999, ApJ, 524, 373)

• “X-ray gap” between ROSAT and OSSE is filled in!

• Paper speculates that:– “Peak 2” may have multiple

components– Some components may be “ γ-

like”, others “optical-like”

• Statistics insufficient to prove either claim

5 April 2001 Gamma2001 Meeting 6

Cycle 3 Provides Some AnswersCycle 3 Provides Some Answers

• Broadband light curves:– First peak clearly aligns

with EGRET first peak– “Second Peak” is clearly

two peaks (∆ϕ=0.09±0.01)• Lower phase, softer

component aligned with optical

• Higher phase, harder component aligned with EGRET second peak

– Soft spectrum peak present at radio peak phase

• Note similar feature in optical lightcurve

• No such feature at higher energies

EGRET> 100 MeV

Pk 1 Pk 2

RXTE16-30 keV

RXTE8-16 keV

RXTE2-8 keV

ROSAT0.06 - 2.4 keV

0.0 0.5 1.0 1.5 2.0Phase

Optical

5 April 2001 Gamma2001 Meeting 7

Cycle 3 Provides Some Answers (cont’d)Cycle 3 Provides Some Answers (cont’d)

• Phase-resolved spectra:– Peaks that are phase-

aligned with γ-ray peaks are hardest

– Peaks phase-aligned with optical and/or radio peak are softest

– Soft component of Peak 2 extrapolates to near optical fluxes

10-2 100 102 104

Energy (keV)

10-6

10-5

10-4

10-3

10-2

Mea

n B

and

En

ergy

x P

hot

on F

lux

(keV

/cm

2 -s-

keV

) RXTE 1st peakRXTE 2nd peak softRXTE 2nd peak hardRXTE radioOSSEROSAT Pulsed BBOptical

5 April 2001 Gamma2001 Meeting 8

Cycle 3 Provides Some Answers (cont’d)Cycle 3 Provides Some Answers (cont’d)

• Total pulsed spectrum– Smoothly fills in gap

between hard X-ray and thermal

– Complex shape from superposition of phase-resolved spectra

– Comparison to Chandra “point source” spectrum indicates higher pulsed fraction than in thermal component (similar to other nonthermal emission)

10-4 10-2 100 102 104 106

Energy (keV)

10-8

10-6

10-4

10-2

Mea

n B

and

En

ergy

x P

hot

on F

lux

(keV

/cm

2 -s-

keV

)RXTE Total PulsedOSSECOMPTELEGRET Total PulsedROSAT Pulsed BlackbodyChandra Point SourceChandra PL EnvelopeOptical

5 April 2001 Gamma2001 Meeting 9

Conclusions and Future DirectionsConclusions and Future Directions

• Conclusions– Cycle 3 RXTE

observations confirm the multicomponent nature of the pulsed Vela emission

– Features from radio through γ-ray present in the 2-30 keV spectrum (hence the Vela Pulsar Rosetta Stone)

– Components behavior in spectrum and phase correlated

• Soft components phase-aligned with lower energy features (radio and optical)

• Hard components phase-aligned with higher energy features (γ-ray)

5 April 2001 Gamma2001 Meeting 10

Conclusions and Future Directions (cont’d)Conclusions and Future Directions (cont’d)

• Conclusions (cont’d)– Nonthermal X-ray

measurements present new challenges to modeling the global spectrum

– Attempts have been made to model these energies (e.g. ⇒ )

– Nothing yet predicts phase resolved spectra behavior

Adapted from Dyks, Rudak & Bulik (00)

5 April 2001 Gamma2001 Meeting 11

Conclusions and Future Directions (cont’d)Conclusions and Future Directions (cont’d)

• Future Directions– More observations with

RXTE (>1000ksec required; not currently approved)

• Significantly improve statistics for characterization of peaks

• Better high energy behavior

– Chandra observations (Pavlov et al)

• Imaging allows phase resolved spectra without subtracting “off-pulse”

• Need better statistics

• Simultaneous Radio– We have full pulse-by-

pulse radio coverage during the cycle 3 observation

– Epoch fold X-ray data for different ranges of radio pulse strength/arrival time

– Study radio phase peak as a function of radio properties